witty-diagnosis-agent 0.3.1-beta → 0.5.0-beta

package/README.md

@@ -9,6 +9,10 @@ Witty 智能诊断 Agent 基于「假设-验证」（Hypothetico-Deductive）故
 - **端到端闭环自愈**：涵盖“现象分析、根因定界、生成报告、执行修复”全链路，提供一站式自动化运维。
 - **严格的安全管控**：诊断排查阶段严格只读，仅修复阶段按需赋予操作权限，且变更必经用户审批确认。
 
+<div align="center">
+  <img src="docs/assets/sys_crash_diag.gif" alt="系统宕机故障诊断" />
+</div>
+
 ### 架构与核心能力
 
 Witty 智能诊断 Agent 采用“Agent-Skill-工具-知识”四层解耦架构，兼具高灵活性与可扩展性。
@@ -43,11 +47,13 @@ Witty 智能诊断 Agent 采用“Agent-Skill-工具-知识”四层解耦架构
 
 > 💡 **更多基于 OS 全栈（硬件层 -> 内核层 -> 系统服务层）的诊断能力持续演进中...**
 
+---
+
 ### 快速开始
 
 #### 环境要求
 
-- 运行环境：Node.js (>=18.0.0)
+- 运行环境：Node.js (>=20.0.0)
 - 依赖工具：已安装[OpenCode](https://opencode.ai/)
 - 依赖工具：已安装 [Ansible](https://docs.ansible.com/ansible/latest/installation_guide/)（`ansible` 命令须在系统 PATH 中可用，可通过 `ansible --version` 提前验证）
 
@@ -74,6 +80,8 @@ cd witty-diagnosis-agent
 bash install.sh
 ```
 
+---
+
 ### 如何使用
 
 1. 启动 \*\*OpenCode \*\*。
@@ -96,7 +104,116 @@ bash install.sh
    
    <img src="docs/assets/guide_auto_diag_report.png" alt="诊断报告" width="800" />
 
-如需了解更多操作细节，可查阅[用户手册](docs/guide/MANUAL.html)；支持的诊断能力与功能说明详见[功能列表](docs/reference/features.html)。
+如需了解更多操作细节，可查阅[用户手册](docs/guide/MANUAL.md)；支持的诊断能力与功能说明详见[功能列表](docs/reference/features.md)。
+
+---
+
+### 进程 CORE 故障诊断示例
+
+本节以 WSL2 Ubuntu 系统为例，演示如何使用 Witty 智能诊断 Agent 分析进程 Core Dump 文件。本示例仅用于演示和体验智能诊断 Agent 的分析过程和输出报告。由于示例代码非常简单（空指针解引用），直接使用原生 OpenCode 或任何调试工具也能定位出根因。对于更复杂的生产环境故障，智能诊断 Agent 的系统化排查能力将展现更大优势。
+
+#### 1. 修改系统配置以获得进程 Core Dump 文件
+
+默认情况下，Linux 系统可能禁用了进程 core dump 功能。请使用 root 用户执行以下步骤启用：
+
+```bash
+# 查看当前 core dump 配置（0 表示禁用）
+ulimit -c
+
+# 临时启用 core dump（不限制文件大小）
+ulimit -c unlimited
+
+# 永久启用：在 /etc/security/limits.conf 末尾添加
+echo "* soft core unlimited" >> /etc/security/limits.conf
+echo "* hard core unlimited" >> /etc/security/limits.conf
+
+# 停用 apport（否则会拦截并吃掉 core 文件）
+sudo service apport stop
+sudo systemctl disable apport
+
+# 设置 core dump 文件存放路径和命名格式（当前目录）
+sudo sysctl -w kernel.core_pattern=core.%e.%p
+
+# 验证配置
+ulimit -c
+cat /proc/sys/kernel/core_pattern
+```
+
+> **说明**：`%e` 为进程名，`%p` 为进程 PID。core 文件将保存在当前工作目录。
+
+#### 2. 开发会产生 Core Dump 的示例 C 代码
+
+创建 `crash_demo.c` 文件：
+
+```c
+#include <stdio.h>
+#include <stdlib.h>
+
+void trigger_segfault() {
+    int *ptr = NULL;
+    *ptr = 42;
+}
+
+int main() {
+    printf("Starting crash demo...\n");
+    trigger_segfault();
+    return 0;
+}
+```
+
+**安装编译工具和 GDB**：
+
+```bash
+sudo apt update && sudo apt install -y gcc gdb
+```
+
+**编译（禁用优化，开启调试符号）**：
+
+```bash
+gcc -g -O0 -o crash_demo crash_demo.c
+```
+
+**运行**：
+
+```bash
+./crash_demo
+```
+
+执行后将触发段错误并产生 core dump 文件：
+
+```
+Starting crash demo...
+Segmentation fault (core dumped)
+```
+
+**查看生成的 core 文件**：
+
+```bash
+ls -la ./core.*
+```
+
+#### 3. 在 WSL 中启动 OpenCode 并使用智能诊断 Agent 分析
+
+**启动 OpenCode**：
+
+```bash
+opencode
+```
+
+**加载 Xuanyuan Agent**：执行 `/agents` 命令，选择 `Xuanyuan` Agent。
+
+**输入问题描述**：
+
+```
+分析/tmp/test目录下的core文件根因。
+```
+
+> **提示**：将 `/tmp/test` 替换为示例 C 代码所在的实际目录路径。
+
+系统将自动执行智能诊断流程，分析进程崩溃原因并输出诊断报告。
+
+---
+
 
 ### 如何贡献
 
@@ -111,6 +228,8 @@ bash install.sh
 - 提交 [issue](https://atomgit.com/openeuler/witty-diagnosis-agent/issues)
 - 发送邮件至 <intelligence@openeuler.org>
 
+---
+
 ### License
 
 本项目中部分代码基于 `oh-my-openagent` 修改，这部分代码遵循 **Sustainable Use License Version 1.0**。其他原创及贡献部分代码协议以最终发布版本为准。详情请参阅项目根目录下的 [LICENSE](./LICENSE) 文件。

package/dist/{auto-update-checker-SCZAPT6L.js → auto-update-checker-FE7T2BSB.js}

@@ -7,7 +7,7 @@ import {
   isDistTag,
   isPrereleaseOrDistTag,
   isPrereleaseVersion
-} from "./chunk-VGS3BNP5.js";
+} from "./chunk-4QO2HAQQ.js";
 import "./chunk-YVNTYBPS.js";
 export {
   checkForUpdate,

package/dist/{chunk-YPQZVQVM.js → chunk-4DUKC3QP.js}

@@ -8,7 +8,7 @@ import {
   migrateConfigFile,
   parseFrontmatter,
   parseJsonc
-} from "./chunk-VGS3BNP5.js";
+} from "./chunk-4QO2HAQQ.js";
 import {
   __require,
   log
@@ -505,6 +505,8 @@ var WebsearchConfigSchema = z24.object({
 // src/config/schema/witty-diagnosis-agent-config.ts
 var WittyDiagnosisAgentConfigSchema = z25.object({
   $schema: z25.string().optional(),
+  /** Global output language for reasoning, interactions, and reports (default: "zh") */
+  output_language: z25.enum(["zh", "en"]).default("zh"),
   /** Enable new task system (default: false) */
   new_task_system_enabled: z25.boolean().optional(),
   /** Default agent name for `witty-diagnosis-agent run` (env: OPENCODE_DEFAULT_AGENT) */

package/dist/{chunk-VGS3BNP5.js → chunk-4QO2HAQQ.js}

@@ -4518,6 +4518,7 @@ function generateModelConfig(config) {
   if (!hasAnyProvider) {
     return {
       $schema: SCHEMA_URL,
+      output_language: config.outputLanguage ?? "zh",
       agents: Object.fromEntries(
         Object.entries(CLI_AGENT_MODEL_REQUIREMENTS).filter(([role, req]) => !(role === "sisyphus" && req.requiresAnyModel)).map(([role]) => [role, { model: ULTIMATE_FALLBACK }])
       ),
@@ -4583,6 +4584,7 @@ function generateModelConfig(config) {
   }
   return {
     $schema: SCHEMA_URL,
+    output_language: config.outputLanguage ?? "zh",
     agents,
     categories
   };
@@ -4645,7 +4647,7 @@ function writeWittyConfig(installConfig) {
           writeFileSync6(wittyConfigPath, JSON.stringify(newConfig, null, 2) + "\n");
           return { success: true, configPath: wittyConfigPath };
         }
-        const merged = deepMergeRecord(newConfig, existing);
+        const merged = deepMergeRecord(existing, newConfig);
         writeFileSync6(wittyConfigPath, JSON.stringify(merged, null, 2) + "\n");
       } catch (parseErr) {
         if (parseErr instanceof SyntaxError) {
@@ -4955,22 +4957,23 @@ import { existsSync as existsSync21, readFileSync as readFileSync19 } from "fs";
 function detectProvidersFromWittyConfig() {
   const wittyConfigPath = getWittyConfigPath();
   if (!existsSync21(wittyConfigPath)) {
-    return { hasOpenAI: true, hasOpencodeZen: true, hasZaiCodingPlan: false, hasKimiForCoding: false };
+    return { hasOpenAI: true, hasOpencodeZen: true, hasZaiCodingPlan: false, hasKimiForCoding: false, outputLanguage: "zh" };
   }
   try {
     const content = readFileSync19(wittyConfigPath, "utf-8");
     const wittyConfig = parseJsonc(content);
     if (!wittyConfig || typeof wittyConfig !== "object") {
-      return { hasOpenAI: true, hasOpencodeZen: true, hasZaiCodingPlan: false, hasKimiForCoding: false };
+      return { hasOpenAI: true, hasOpencodeZen: true, hasZaiCodingPlan: false, hasKimiForCoding: false, outputLanguage: "zh" };
     }
     const configStr = JSON.stringify(wittyConfig);
     const hasOpenAI = configStr.includes('"openai/');
     const hasOpencodeZen = configStr.includes('"opencode/');
     const hasZaiCodingPlan = configStr.includes('"zai-coding-plan/');
     const hasKimiForCoding = configStr.includes('"kimi-for-coding/');
-    return { hasOpenAI, hasOpencodeZen, hasZaiCodingPlan, hasKimiForCoding };
+    const outputLanguage = wittyConfig.output_language ?? "zh";
+    return { hasOpenAI, hasOpencodeZen, hasZaiCodingPlan, hasKimiForCoding, outputLanguage };
   } catch {
-    return { hasOpenAI: true, hasOpencodeZen: true, hasZaiCodingPlan: false, hasKimiForCoding: false };
+    return { hasOpenAI: true, hasOpencodeZen: true, hasZaiCodingPlan: false, hasKimiForCoding: false, outputLanguage: "zh" };
   }
 }
 function detectCurrentConfig() {
@@ -4983,7 +4986,8 @@ function detectCurrentConfig() {
     hasCopilot: false,
     hasOpencodeZen: true,
     hasZaiCodingPlan: false,
-    hasKimiForCoding: false
+    hasKimiForCoding: false,
+    outputLanguage: "zh"
   };
   const { format, path: path9 } = detectConfigFormat();
   if (format === "none") {
@@ -5000,11 +5004,12 @@ function detectCurrentConfig() {
     return result;
   }
   result.hasGemini = plugins.some((p) => p.startsWith("opencode-antigravity-auth"));
-  const { hasOpenAI, hasOpencodeZen, hasZaiCodingPlan, hasKimiForCoding } = detectProvidersFromWittyConfig();
+  const { hasOpenAI, hasOpencodeZen, hasZaiCodingPlan, hasKimiForCoding, outputLanguage } = detectProvidersFromWittyConfig();
   result.hasOpenAI = hasOpenAI;
   result.hasOpencodeZen = hasOpencodeZen;
   result.hasZaiCodingPlan = hasZaiCodingPlan;
   result.hasKimiForCoding = hasKimiForCoding;
+  result.outputLanguage = outputLanguage;
   return result;
 }
 

package/dist/cli.js

@@ -13,7 +13,7 @@ import {
   readBoulderState,
   readContinuationMarker,
   readState
-} from "./chunk-YPQZVQVM.js";
+} from "./chunk-4DUKC3QP.js";
 import {
   AGENT_MODEL_REQUIREMENTS,
   CATEGORY_MODEL_REQUIREMENTS,
@@ -42,7 +42,7 @@ import {
   parseJsonc,
   spawnWithWindowsHide,
   writeWittyConfig
-} from "./chunk-VGS3BNP5.js";
+} from "./chunk-4QO2HAQQ.js";
 import "./chunk-YVNTYBPS.js";
 
 // src/cli/cli-program.ts
@@ -51,7 +51,7 @@ import { Command as Command2 } from "commander";
 // package.json
 var package_default = {
   name: "witty-diagnosis-agent",
-  version: "0.3.1-beta",
+  version: "0.5.0-beta",
   description: "Witty Diagnosis Agent - OpenCode Plugin",
   keywords: [
     "witty-diagnosis-agent"
@@ -62,6 +62,9 @@ var package_default = {
   },
   license: "MIT",
   author: "witty-diagnosis-agent",
+  engines: {
+    node: ">=20.0.0"
+  },
   type: "module",
   main: "dist/index.js",
   exports: {
@@ -240,6 +243,9 @@ function validateNonTuiArgs(args) {
   if (args.kimiForCoding !== void 0 && !["no", "yes"].includes(args.kimiForCoding)) {
     errors.push(`Invalid --kimi-for-coding value: ${args.kimiForCoding} (expected: no, yes)`);
   }
+  if (args.outputLanguage !== void 0 && !["zh", "en"].includes(args.outputLanguage)) {
+    errors.push(`Invalid --output-language value: ${args.outputLanguage} (expected: zh, en)`);
+  }
   return { valid: errors.length === 0, errors };
 }
 function argsToConfig(args) {
@@ -251,7 +257,8 @@ function argsToConfig(args) {
     hasCopilot: args.copilot === "yes",
     hasOpencodeZen: args.opencodeZen === "yes",
     hasZaiCodingPlan: args.zaiCodingPlan === "yes",
-    hasKimiForCoding: args.kimiForCoding === "yes"
+    hasKimiForCoding: args.kimiForCoding === "yes",
+    outputLanguage: args.outputLanguage ?? "zh"
   };
 }
 function detectedToInitialValues(detected) {
@@ -266,7 +273,8 @@ function detectedToInitialValues(detected) {
     copilot: detected.hasCopilot ? "yes" : "no",
     opencodeZen: detected.hasOpencodeZen ? "yes" : "no",
     zaiCodingPlan: detected.hasZaiCodingPlan ? "yes" : "no",
-    kimiForCoding: detected.hasKimiForCoding ? "yes" : "no"
+    kimiForCoding: detected.hasKimiForCoding ? "yes" : "no",
+    outputLanguage: detected.outputLanguage ?? "zh"
   };
 }
 async function checkAnsibleInstalled() {
@@ -453,8 +461,21 @@ import * as p from "@clack/prompts";
 import color3 from "picocolors";
 
 // src/cli/tui-install-prompts.ts
+import { select, isCancel } from "@clack/prompts";
 async function promptInstallConfig(detected) {
   const initial = detectedToInitialValues(detected);
+  const languageResult = await select({
+    message: "Select output language for Witty Diagnosis Agent / \u9009\u62E9 Witty Diagnosis Agent \u7684\u8F93\u51FA\u8BED\u8A00\n  (This affects all agents' reasoning, response, and report / \u5F71\u54CD\u6240\u6709 Agent \u7684\u601D\u8003\u3001\u56DE\u590D\u548C\u751F\u6210\u7684\u62A5\u544A)",
+    options: [
+      { value: "zh", label: "\u7B80\u4F53\u4E2D\u6587 (zh)" },
+      { value: "en", label: "English (en)" }
+    ],
+    initialValue: initial.outputLanguage ?? "zh"
+  });
+  if (isCancel(languageResult)) {
+    return null;
+  }
+  const outputLanguage = languageResult;
   const claude = initial.claude ?? "no";
   const openai = initial.openai ?? "no";
   const gemini = initial.gemini ?? "no";
@@ -470,7 +491,8 @@ async function promptInstallConfig(detected) {
     hasCopilot: copilot === "yes",
     hasOpencodeZen: opencodeZen === "yes",
     hasZaiCodingPlan: zaiCodingPlan === "yes",
-    hasKimiForCoding: kimiForCoding === "yes"
+    hasKimiForCoding: kimiForCoding === "yes",
+    outputLanguage
   };
 }
 
@@ -2191,7 +2213,7 @@ async function getLocalVersion(options = {}) {
       console.log(options.json ? formatJsonOutput(info2) : formatVersionOutput(info2));
       return 1;
     }
-    const { extractChannel: extractChannel2 } = await import("./auto-update-checker-SCZAPT6L.js");
+    const { extractChannel: extractChannel2 } = await import("./auto-update-checker-FE7T2BSB.js");
     const channel = extractChannel2(pluginInfo?.pinnedVersion ?? currentVersion);
     const latestVersion = await getLatestVersion(channel);
     if (!latestVersion) {
@@ -3705,11 +3727,11 @@ function createMcpOAuthCommand() {
 var VERSION2 = package_default.version;
 var program = new Command2();
 program.name("witty-diagnosis-agent").description("The ultimate OpenCode plugin - multi-model orchestration, LSP tools, and more").version(VERSION2, "-v, --version", "Show version number").enablePositionalOptions();
-program.command("install").description("Install and configure witty-diagnosis-agent with interactive setup").option("--no-tui", "Run in non-interactive mode (requires all options)").option("--claude <value>", "Claude subscription: no, yes, max20").option("--openai <value>", "OpenAI/ChatGPT subscription: no, yes (default: no)").option("--gemini <value>", "Gemini integration: no, yes").option("--copilot <value>", "GitHub Copilot subscription: no, yes").option("--opencode-zen <value>", "OpenCode Zen access: no, yes (default: no)").option("--zai-coding-plan <value>", "Z.ai Coding Plan subscription: no, yes (default: no)").option("--kimi-for-coding <value>", "Kimi For Coding subscription: no, yes (default: no)").option("--skip-auth", "Skip authentication setup hints").addHelpText("after", `
+program.command("install").description("Install and configure witty-diagnosis-agent with interactive setup").option("--no-tui", "Run in non-interactive mode (requires all options)").option("--claude <value>", "Claude subscription: no, yes, max20").option("--openai <value>", "OpenAI/ChatGPT subscription: no, yes (default: no)").option("--gemini <value>", "Gemini integration: no, yes").option("--copilot <value>", "GitHub Copilot subscription: no, yes").option("--opencode-zen <value>", "OpenCode Zen access: no, yes (default: no)").option("--zai-coding-plan <value>", "Z.ai Coding Plan subscription: no, yes (default: no)").option("--kimi-for-coding <value>", "Kimi For Coding subscription: no, yes (default: no)").option("--output-language <value>", "Output language for agents: zh, en (default: zh)").option("--skip-auth", "Skip authentication setup hints").addHelpText("after", `
 Examples:
   $ npx witty-diagnosis-agent install
   $ npx witty-diagnosis-agent install --no-tui --claude=max20 --openai=yes --gemini=yes --copilot=no
-  $ npx witty-diagnosis-agent install --no-tui --claude=no --gemini=no --copilot=yes --opencode-zen=yes
+  $ npx witty-diagnosis-agent install --no-tui --claude=no --gemini=no --copilot=yes --opencode-zen=yes --output-language=en
 
 Model Providers (Priority: Native > Copilot > OpenCode Zen > Z.ai > Kimi):
   Claude        Native anthropic/ models (Opus, Sonnet, Haiku)
@@ -3729,6 +3751,7 @@ Model Providers (Priority: Native > Copilot > OpenCode Zen > Z.ai > Kimi):
     opencodeZen: options.opencodeZen,
     zaiCodingPlan: options.zaiCodingPlan,
     kimiForCoding: options.kimiForCoding,
+    outputLanguage: options.outputLanguage,
     skipAuth: options.skipAuth ?? false
   };
   const exitCode = await install(args);

package/dist/index.d.ts

@@ -1093,6 +1093,10 @@ type HookName = z.infer<typeof HookNameSchema>;
 
 declare const WittyDiagnosisAgentConfigSchema: z.ZodObject<{
     $schema: z.ZodOptional<z.ZodString>;
+    output_language: z.ZodDefault<z.ZodEnum<{
+        zh: "zh";
+        en: "en";
+    }>>;
     new_task_system_enabled: z.ZodOptional<z.ZodBoolean>;
     default_run_agent: z.ZodOptional<z.ZodString>;
     disabled_mcps: z.ZodOptional<z.ZodArray<z.ZodString>>;